The proposed research will improve the capability for determining high- resolution protein structures in solution from two-dimensional NMR experiments. Part of this advance will come via implementation of methodology developed in this lab for the complete relaxation matrix analysis (CORMA) of two-dimensional nuclear Overhauser effect (2D NOE) spectra, enabling the determination of a large number of internuclear distances at considerably greater accuracy than previously possible. The new methodology should also enable somewhat longer distances to be determined-up to 5angstrom or possibly 6angstrom. The most effective means (as well as the influence of plausible errors and assumptions) of deriving protein solution structure by using the complete relaxation matrix analysis in conjunction with the computational techniques of distance geometry, molecular mechanics and restrained molecular dynamics will be explored. The distance information obtained via 2D NOE spectra will be augmented by dihedral angle constraints derived from scalar coupling-based 2D NMR experiments. The effects of molecular motions and multiple conformations will also be examined. Programs of merit which are developed by this project will be disseminated to other research labs. The improved methodology will be especially useful in protein moieties where improved structural knowledge is highly desirable, i.e., in ligand-binding sites. The improved structures should lead to greater mechanistic insights.